SUMMARYThe widespread conviction that perceiving another person must rest on ambiguous and fakeable information is challenged. Arguing from biomechanical necessities inherent in maintaining balance and coping with reactive impulses, we show that the detailed kinematic pattern is specific to an acting person's anatomical makeup and to the working of his or her motor control system. In this way information is potentially available about gender, identity, expectations, intentions, and what the person is in fact doing. We invoke the lawfulness of human movement, as elucidated by recent advances in motor control theory, to demonstrate the virtual impossibility of performing truly deceptive movements and to argue in general terms for the specification power inherent in human kinematics.The outcome of the analysis is subsumed under a principle of kinematic specification of dynamics (KSD), which states that movements specify the causal factors of events. Generally, a linked multiple degrees-of-freedom system does not exhibit substitutability; a change in one of its "input" factors cannot substitute for, or cancel, the multivariable effects of a change in another factor.Six explorative experiments are reported. Displaying humans in action with Johansson's (1973) patch-light technique, we show that (a) the influence of an invisible thrown object on the kinematics of the thrower enable observers to perceive the length of throw; (b) the lead-in movements of a person lifting a box allow perception of what weight the lifter expects; (c) a person lifting a box cannot deceive observers about the weight of the box, only convey the deceptive intention; (d) gender of adults and children in complex activity is recognizable to about 75% of presentations; (e) gender recognition rises to about 85% correct when the observed persons are.not self-conscious about gender; (f) real gender and expressed (acted) gender are simultaneously, but independently, perceived; and (g) observer instructions to judge only "gender" yields results that erroneously indicate a deception effect.We conclude that the experiments have demonstrated the considerable effectiveness of kinematic information in enabling perception of persons and action. Judgments of good precision were often obtained. Some perceived properties were relatively subtle states of the seen person. True conditions were perceived despite deceptive endeavors. The KSD principle therefore appears an appropriate conceptual guide, and the patch-light technique a useful empirical method, for the study of social knowing.The concluding discussion argues that person perception has a dual nature in that true person properties and communicative or deceptive expressions are co-specified in the kinematic pattern. Hence, they constitute alternative foci for perception, and attention can switch freely between them in normal social interaction. Extensive parallels exist between this view of person perception and Gibson's (1979) treatment of pictures and picture perception.Furthermore, we argue that the stu...

The weight of a box can be seen by observing another person lifting and carrying it. Evidence is provided in two experiments, the first of which employed videotaped events with the actor and box visible only as 21 bright patches. Observers judged the weight of the box rather linearly with an average slope of .87 and with a pooled standard deviation of 3.8 kg. The second experiment compared visual and haptic perception of box weight in similar events under conditions of live action. Average slopes of 1.00 in the visual mode and 1.20 in the haptic mode were obtained with standard deviations of 3.1 kg and 2.0 kg, respectively. It is concluded that the weight of the box, as a dynamic variable of the event, is well specified in the kinematic pattern and hence in the optic array. Furthermore, the visual system is efficient in picking up such information.

Constructivist and Gibsonian approaches disagree over the possibility of direct perceptual use of advanced information. A trenchant instance concerns visual perception of underlying dynamic properties as specified by kinematic patterns of events. For the paradigmatic task of discrimination of relative mass in observed collisions, 2 mathematical models are developed, 1 model representing a direct, invariant-based approach, and 1 representing a cue-heuristic approach. The models enable a critical experimental design with distinct predictions concerning performance data and confidence ratings. Although pretraining results were mixed, the invariant-based model was empirically confirmed after a minimal amount of training: Competence entails the use of advanced kinematic information in a direct-perceptual ("sensory") mode of apprehension, in contrast to beginners' use of simpler cues in an inferential ("cognitive") mode.

A basic feature of some modem theories of perception is the notion of complex or higher order variables which are considered basic for perception. A distinction between "rote" and "smart" mechanisms is introduced and it is suggested that perception consists of smart mechanisms which directly register complex variables. A model of a perceiver, based on the smart polar planimeter, is constructed and used to illustrate the possible consequences of smart perceptual mechanisms for research in areas such as psychophysics, cognition, attention, and perceptual development and learning. 15, 17-34.

Novice observers differ from each other in the kinematic variables they use for the perception of kinetic properties, but they converge on more useful variables after practice with feedback. The colliding-balls paradigm was used to investigate how the convergence depends on the relations between the candidate variables and the to-be-perceived property, relative mass. Experiment 1 showed that observers do not change in the variables they use if the variables with which they start allow accurate performance. Experiment 2 showed that, at least for some observers, convergence can be facilitated by reducing the correlations between commonly used nonspecifying variables and relative mass but not by keeping those variables constant. Experiments 3a and 3b further demonstrated that observers learn not to rely on a particular nonspecifying variable if the correlation between that variable and relative mass is reduced.